Title :
A Kind of Improved AdaBoost Algorithm
Author :
Liao Shaowen ; Chen Yong
Author_Institution :
Inf. Technol. & Media Inst., Hexi Univ., Zhangye, China
Abstract :
AdaBoost algorithm is a kind of very important feature classification machine learning algorithm, But if difficult samples exist in the training samples, with the iterative Nurnber increasing, this easily leads to degeneration Phenomenon, and reduces the generalization ability of the classifier. In view of the face detection under complex background degeneration appeared problem, This article Proposes LWE-AdaBoost algorithm which can limit weight expansion, the experimental results indicate that the LWE-AdaBoost algorithm can restrain the recurrence of degeneration Phenomenon well.
Keywords :
learning (artificial intelligence); pattern classification; LWE-AdaBoost algorithm; complex background degeneration; face detection; feature classification machine learning algorithm; generalization ability; iterative number; training; Algorithm design and analysis; Classification algorithms; Face; Face detection; Machine learning algorithms; Standards; Training; AdaBoost algorithm; Cascaded classifier; Face detection; Integral image; Strong classifier; Weak classifier;
Conference_Titel :
Intelligent Computation Technology and Automation (ICICTA), 2014 7th International Conference on
Conference_Location :
Changsha
Print_ISBN :
978-1-4799-6635-6
DOI :
10.1109/ICICTA.2014.11